NNEF 1.0 Standard

After gathering feedback from the provisional specification, NNEF 1.0 has been released as a stable, flexible, and extensible open standard for hardware manufacturers to reliably enable widespread deployment of neural networks onto edge devices. New tools are now available on Github including an NNEF parser and converters from Tensorflow and Caffe. There are importers too into popular inferencing environments including Android’s Neural Network API under development.

NNEF was created to reduce fragmentation by facilitating the exchange of neural networks among training frameworks and inference engines, increasing the freedom for users to mix and match the inference engines and training frameworks of their choice. The standard, the result of collaboration between industry-leading members of the  Khronos Working Group, guarantees the stability that hardware and software companies can rely on for product deployment while maintaining the flexibility required to respond to the needs of rapidly-moving industries.

01 – NNEF is the latest new standard from the Khronos Group and deals with neural networks.

NNEF 1.0 will allow reliable export from the rapidly growing number of frameworks, including Torch, Chainer, Theano, PyTorch, and MXNet in addition to those for which tools already exist. A complete workflow from training, through optimization, to deployment is possible using NNEF as the standard transfer format between tools. Open source tools are available to allow easy creation of importers into custom and embedded edge-inferencing environments. For example, open source efforts are already underway to produce importers into more general-purpose environments such as NN API and CoreML.

NNEF was designed to accommodate a wide range of use-cases and network types. Its basic format is designed to be both human readable and easy to parse, whereas its extensions allow for human editing of large networks. It is designed to accommodate rapid evolution of frameworks without the need to change the base standard, and provides an extension mechanism to handle specific issues such as custom data formats for trained network weights.

At launch, the standard will be supported by two open source Tensorflow converters both for network descriptions based on protobuf and python code, and as a converter for Caffe. A Caffe2 open-source converter in development by Au-Zone Technologies will be available in Q3 2018. Various tools from member companies Almotive and AMD will be available, and an importer to the Android NN API is under development by a team from National Tsing-Hua University of Taiwan.

“Khronos recognized a growing format compatibility problem for companies wanting to deploy trained neural networks onto edge devices. We set out to build the first standard platform for engineers to easily deploy networks from training frameworks to inference engines. Today, we are proud to release NNEF 1.0 as a stable, open specification that will allow existing hardware to work but continue to evolve through flexible extension mechanisms to keep up with this fast changing field of machine learning,” said Khronos NNEF Working Group Chair Peter McGuinness. “In December 2017 we released the developer preview of NNEF and made an open call for industry feedback. Community response has been tremendous, confirming the demand for this standard and enabling us to achieve a responsive and complete NNEF 1.0 specification.”

The NNEF 1.0 specification and documentation are freely available on the Khronos website, and NNEF open source tools and projects are available in the Khronos NNEF Tools repository.

The post Khronos Group Releases Final NNEF 1.0 Standard For Deployment of Trained Neural Networks appeared first on Architosh.

Their end goal is to empower data scientists and engineers to facilitate the exchange of trained networks from chosen training framework to a whole big variety of inference engines. The new format is versatile and standard to an extent that equipment manufacturers can depend on its deployment onto edge devices. NNEF captures the complete description of structure and parameters of the neural network, for independent training tools, and interference engine used for the execution.

Peter McGuinness, Chairperson at NNEF work, says, “The field of machine learning benefits from the vitality of the many groups working in the field, but suffers from a lack of common standards, especially as research moves closer to multiple deployed systems.” He further adds that “Khronos anticipated this industry need and has been working for over a year on the NNEF universal standard for neural network exchange, which will act as the equivalent of a pdf for neural networks.”

01 – The Khronos Group has established NNEF to help reduce machine learning deployment fragmentation and enable rich mix of tools for the industry. (image: Khronos Group / Architosh. All rights reserved.)

Torch, Caffe, and Caffe2, TensorFlow, Theano, Chainer, MXNeT are a few of the many tools and engines across which NNEF can perform the exchange with reliability. However, NNEF 1.0 Provisional specification encompasses a bunch of use cases and multiplicity of network types and a scalable design that borrows syntactical elements from Python and aids correctness by adding formal elements. This also includes custom compound operations and posh network optimization. The entire goal is to draw the future work on this very same architecture for providing a stable platform in the field of quickly evolving machine learning.

The giant has triggered a sequence of open source projects, which includes NNEP syntax parser or validator from selected frameworks and encourages machine learning community to use NNEF for their workflows.

Additionally, NNEF is all set to work closely with popular product, OpenVX, of Khronos group for enabling incorporation of NNEF files for deep learning operations in a single graph by leveraging OpenVX 1.2 as a cross-platform interface.

Industry Says

An industry-wide support is apparent as the announcement of the release of NNEF 1.0, and the professionals are taking all the pride that they can during the development of the network. From hardware engineering key players like Almotive to Radeon Technologies group and from Machine learning community Arm, to wireless technology forerunners, Qualcomm technologies, everyone is on the same positive front, as the development of NNEF. As the vice presidents and chairpersons come to the support, there is an industry-wide flare of acceptance and eager wait for flexibility in handling flat and compound operations.

The new NNEF 1.0 documentation project and specification are available on the Khronos Registry. Please visit The Khronos Group official site for more information.

SaveSave

SaveSave

SaveSave

SaveSave

The post Khronos Group—Spells Out Details of Neural Network Standard appeared first on Architosh.

One can also take geographic analysis further with MapBox integration, a leading map platform that lets you drop in satellite imagery with a few clicks, or incorporate custom maps. They are also bringing a slew of new enterprise features such as project level permissions, updated connectors and new authentication technologies.

Mobile

Tableau is on the iPhone, to download the new iOS app click here. Tableau Mobile is the fastest, most delightful way to stay on top of your data on your tablet or phone. Use the native app to see and understand data with only a few taps. Tableau is also location-aware and when being off the grid location can be found and even get offline snapshots of favorite vizzes. Tableau Mobile also lets users find content faster with Faceted Search.

Mapping

The new mapping data and formatting options allows users to simply dive deeper into their geographical analysis. Integration with Mapbox gives the ability to build gorgeous map background and add layers and contexts.

01 – Customize how people interact with your map with the new Map Options Dialog. Change display options for pan and zoom, map search and the view toolbar.

Tableau 9.2 comes with more geographical data built in, including updated US congressional districts, local name synonyms for world capitals, Japanese postal codes and more.

Visual Analytic

Questions of data can be answered with expanded visual analytics capabilities. Treemaps now show hierarchical data, by introducing borders around the first level of grouping.

02 – Put totals and subtotals where you want them: top or bottom, left or right.

Visual analytic has Label Options that matches label colors to mark colors for at-a-glance understanding and specify which marks are labeled with the new “Most Recent” option. Further capabilities are Web Authoring, Fix Axis, Recent Colors and the ability to use sheets as filters.

Data prep

Big time data clean up in a single click. Automatically detect sub-tables and covert them to connection-friendly tables with Tableau’s Data Interpreter. Data Grid Editing allows to sort columns and rows, edit column names and create calculations directly in the data grid while Data Pane quickly searches and changes a field’s data type with the updated data pane.

Enterprise

Enjoy new authentications, performance and governance features such as Set Project Permissions, allowing for more flexible configurations, Lock Project Permissions allows administrators and project leaders to lock a project’s permissions. Enterprise also contains Toolbar Improvements, performance and authentication, where users can automatically sign in to Tableau Server as long as they are signed in to an OpenID Connect identity provider such as Google.

Try Tableau 9.2 by downloading a fully functional free trial. Accelerate and empower your skills with hands-on educations. Tableau’s expert trainers will guide you through intensive courses that can help you get the most out of Tableau’s software. For upcoming courses click, here.

The post The new Tableau 9.2—Your Data is Calling appeared first on Architosh.

These interactive exercises are engaging in different degrees of complexity by combining education and entertainment that identifies strengths and learning gaps.

Pixar in a Box

Pixar in a Box offers instructional online video lessons to empower learners while letting them study at their own pace, outside of the classroom.

01 – Watch the video of Pixar in a Box overview.

Participants can learn along the way a lot about Pixar’s filmmaking process and how Pixar artists use math to solve design problems during the stages of the creative process by particularly focusing on specific areas like parabolic arcs, weighted averages and animation curves.

Pixar production pipeline topics

There are already over 100,000 interactive exercises that can sharpen users’ skills from math, science through computer programming, history, art and economics with Khan Academy; and while these lessons are in the first phases of this project, with the first year primarily focusing on math, it’s worth noting that future lessons will target the above topics. Each topic begins with a design-focused lesson, followed by a math-focused lesson.

Topic overview

• Enviroment Modeling – explore how realistic blades of grass are modeled using parabolic arcs
• Character Modeling – explore how clay models are transformed into digital characters using weighted averages
• Building Crowds – explore how swarm of robots were created for WALL-E using combinatorics
• Animation – explore how animators bring characters to life with the help of Animation curves
• Sets & Staging – explore how virtual sets are constructed using geometric transformations
• Rendering – explore how pixels are painted by solving systems of linear equations

The lessons concepts are ranging from grade 4 through high school and beyond. These tools are free for all ages, from teens to parents and teachers. Khan Academy offers practice exercises, instructional videos and a personalized learning dashboard with a mission to guide learners from kindergarten to calculus using state-of-the-art, adaptive technology.

The post Disney News: ‘Pixar in a Box’ launches free for all and forever online programs appeared first on Architosh.

Targeting K-12 schools

Project Ignite encompasses the end-to-end design experience, from idea to digital modeling to physical fabrication. It brings together free* and custom design software, step-by-step projects and for-purchase hardware options. Autodesk’s Project Ignite, Tinkercad and 123D Circuits web sites are COPPA safe-harbor certified by CARU Kid’s Privacy Safe Harbor Program.

Education offering

With Autodesk’s long history in education, it understands first-hand how young learners have the potential to change the world when given the right tools and inspiration. The Project Ignite learning platform adds to Autodesk’s comprehensive free* offerings in education by delivering a unique package of technology, learning content and services created specifically to bring the Maker Movement into the classroom.

01 – Project Ignite is aimed at getting kids interested in innovative futures.

Samir Hanna, vice president and general manager, Autodesk Consumer and 3D Printing said, “The ability to take risks, learn from failure and solve problems is critical to prepare students for the challenges ahead”.

Additional supports

With support from companies including Microsoft, MakerBot, and Electroninks Writeables, the Project Ignite learning platform additionally offers classroom bundles, which include hardware such as 3D printers and electronics kits along with professional development and training services, to help educators get up and running quickly.

Autodesk aims to bring innovative technology on common platforms to offer a better end user experience. Just as Spark, Autodesk’s open 3D printing platform, aims to accelerate innovation in 3D printing, Project Ignite extends this concept into education by empowering the next generation of innovators and will be available in English for educators and students in the U.S. and abroad.

• For educators – Project Ignite brings together free* Autodesk design software, including Tinkercad and 123D Circuits, and offers step-by-step lessons within these apps to encourage design thinking and hands-on problem solving skills. It also provides for-purchase hardware options such as 3D printers from MakerBot and Circuit Scribe pens and modules from Electroninks Writeables. Educators can also login for free* to access Project Ignite content whether or not they wish to purchase hardware.
• For schools – Administrators may choose from various package solutions comprised of free* software and for-purchase hardware options, educator training and support, all from a single place to introduce the Maker Movement into the classroom.
• For educational partners – Project Ignite is open to educators, publishers, education technology and hardware providers looking for unique opportunities to help drive the next education revolution that will inspire students to further their STEAM (Science, Technology, Engineering, Art and Math) education and provide educators access to emerging technology trends quickly.
• For families or parents – Project Ignite is a great way to learn technology basics together in a moderated environment that extends to engaging and fun weekend project

To learn more about the Project Ignite learning platform, visit https://projectignite.autodesk.com.

More on Autodesk in Education

Autodesk provides young learners, educators and schools across the education continuum with access to its portfolio of free* professional 3D design software and creative applications, as well as project-based learning content to help inspire and prepare the next generation of makers, creators and innovators. This includes:

• Building proficiency with young learners from elementary school and up to support STEAM related subjects to solve real world challenges.
• Imparting industry-relevant knowledge and 3D design skills to give students a competitive edge to achieve their personal goals today, and career success in the future.
• Helping educators to inspire creativity and innovation amongst their students, and facilitate a multidisciplinary approach to education that reflect today’s business realities.

To request free access to Autodesk software for schools, please visit www.autodesk.com/academic.

About Autodesk

Autodesk helps people imagine, design and create a better world. Everyone—from design professionals, engineers and architects to digital artists, students and hobbyists—uses Autodesk software to unlock their creativity and solve important challenges. For more information visit autodesk.com or follow @autodesk.

The post Autodesk Project Ignite Brings Innovation to Young Learners appeared first on Architosh.

The New OpenCL 2.1

A significant evolution of the royalty-free standard for heterogeneous parallel programming, the new OpenCL 2.1 spec importantly will now support C++ as a programming language, significantly improving programming productivity. It will also now support Khronos SPIR-V™ a cross-API shader program intermediate language now used by both OpenCL and the new Vulkan API.

OpenCL 2.1 is being released in provisional form to enable developers and implementers to provide feedback before the finalization at the OpenCL forums.

GDC Session

On Wednesday, 4 March 2015, there will be a session on the new OpenCL 2.1 titled, OpenCL Ecosystem Advances: OpenCL 2.1, SPIR-V and SYCL. The event is at 3-4:30 PM and the venue is the SF Green Space at 657 Mission Street, Suite 200, a five minute walk from the GDC (Game Developers Conference) in San Francisco this week.

No GDC pass is required to attend however seating is limited and those wishing to attend should register here.

Big Improvements in OpenCL: The Details

Perhaps the biggest improvement coming to OpenCL in version 2.1 is the C++ language support. OpenCL C++ enables reusable device libraries and containers for easily sharable code that is fast and elegant, and templates enables meta-programming for highly adaptive software that cleanly delivers performance portability.

01 – The new OpenCL 2.1 supports a richer compiler ecosystem, offering the developer and hardware community a superior compute GPU-based API system.  (image: courtesy of Khronos Group, All rights reserved.)

The second key feature is the new shared shader language in SPIR-V support. What SPIR-V does is splits the compiler chain, enabling high-level language front-ends to emit programs in a standardized intermediate form to be ingested by Vulkan or OpenCL drivers. This feature will eliminate the need for a built-in high-level language source compiler and thereby reduces driver complexity and at the same time offers a diversity of possible language front-ends.

New features in OpenCL at the API level include:

• Subgroups, that expose hardware threading, are brought into core, together with additional subgroup query operations for increased flexibility;
• clClonekernel — enables copying of kernel objects and state for safe implementation of copy constructors in wrapper classes’
• Low-latency device timer queries support alignment of profiling between device and host code

 “OpenCL 2.1 has responded to developer demand with a C++ based kernel language which delivers the next level of programmer productivity in parallel programming, while still preserving backwards compatibility for existing OpenCL C kernels,” said Neil Trevett, president of the Khronos Group and chair of the OpenCL working group and vice president at NVIDIA.  “The use of SPIR-V by Vulkan and OpenCL will fundamentally reshape the graphics and compute ecosystem by enabling diverse language and middleware front-ends to leverage the hardware community’s investment in optimized back-end drivers. Khronos is investigating catalyzing the OpenCL 2.1 ecosystem with an open source front-end OpenCL C++ compiler implementation and a convertor between SPIR-V and LLVM, and we welcome developer feedback on this and any other aspect of OpenCL 2.1.”

Industry Support for OpenCL 2.1 — Industry Reaction

“AMD is excited to see OpenCL™ evolve to include a C++ kernel language, which will significantly expand the number of developers targeting heterogeneous platforms,” said Manju Hegde, corporate vice president, Heterogeneous Applications and Developer Solutions, AMD.  “We also applaud the bold move to SPIR-V which provides a common binary target language across graphics and compute.  This greatly simplifies vendor driver development and enables innovation in new languages targeting cross-platform acceleration of applications.”

“Mobile devices are now the primary computing platform for consumers,” said Dennis Laudick, vice president, partner marketing, media processing group, ARM. “As a founding member of Khronos, ARM is fully behind the new OpenCL specification including new features that enable mobile computing to be more energy-efficient.”

 “Intel has been deeply involved in the development of OpenCL 2.1. We’re excited to get this new specification into the hands of Intel platform developers,” said Jon Khazam, vice president and general manager of Intel’s Visual & Parallel Computing Group.  “The addition of C++ support and movement toward a Common IR across OpenCL and the new Vulkan graphics API will facilitate wider adoption and deployment of OpenCL in heterogeneous computing applications.”

“The addition of the C++ kernel language in OpenCL 2.1 is a very exciting development for the HPC community,” said Simon McIntosh-Smith, head of the HPC research group at the University of Bristol. “C++ is being increasingly used to develop scientific codes, and so this development will make it much easier to port new HPC software to a wide variety of high performance, parallel hardware. OpenCL 2.1’s C++ kernel language could cause a step change in the rate of adoption for OpenCL in HPC.”

“We at YetiWare believe that software should run as fast as possible by using all of the processors on a system, and the OpenCL standard makes that possible,” said AJ Guillon, founder and chief technical officer, YetiWare Inc.  “The OpenCL C++ kernel language and SPIR-V are both major advancements and big wins for developers.”

Learning More

The Khronos Group is announcing several things this week at GDC, along with OpenCL 2.1 they have also announced Vulkan™ the new graphics and compute API for GPUs. You can learn more here.

The post GDC: Khronos Releases OpenCL 2.1 Provisional Specification for Public Review appeared first on Architosh.

The Process

This bust, completed earlier this year, was undertaken by the Digitization Program Office of the Smithsonian Institution. Unlike the grease allegedly applied to President Lincoln’s beard, President Obama had a much cleaner and simpler task for sitting for his bust portrait.

Obama needed to tolerate 1 second of flashing lights, as well as sit still for about 90 seconds. A team led by Dr. Paul Debevec of USC’s Institute of Creative Technologies, assembled a Mobile Light Stage which took a total of 80 photographs of President Obama using eight high-end DSLRs and 50 light sources…10 photographs each under different light levels.

01 – Digital portraits of the first digitized 3D presidential bust. (courtesy of Smithsonian Institute-DPO. All rights reserved)

Sophisticated algorithms processed this data into high-resolution 3D models capturing the president’s facial features from ear to ear. Then a hand 3D scanner was used to capture the 3D geometry of the president’s entire head.

Autodesk’s Role

After this work was completed the first batch of data was transferred to Autodesk where the Mobile Light Stage, hand scanned data, and photographic data were all combined and post-processed. The Autodesk team put together a 3D model that combined all the data like a puzzle, precisely registering the geometry from different data sources into a unified 3D space.

After this process color control processes were undertaken. Then a digital modeler sculpted a plinth to complete the bust. After just 72 hours the entire undertaking produced a master file of 15 million triangles, which also generated 3D print-level files for the Obama bust and life mask to be produced in the physical. The data was transferred to 3D Systems who processed the final form in a Selective Laser Sintering (SLS) process that melts nylon powder into a highly accurate and durable 3D print.

Viewing the Obama Bust

The President Obama Digital Bust was unveiled 18 June 2014 at the White House Maker Faire and are now on display in the Smithsonian Castle.

A new video released by the White House (see above) details the behind-the-scene process of scanning, creating, and printing the historic portraits. From 2 December 2014 through 31 December 2014, the portraits will be on view in the Commons gallery of the Smithsonian Castle.

Notes

To learn more about the detailed process involved please read here: http://dpo.si.edu/blog/smithsonian-creates-first-ever-3d-presidential-portrait

The post Smithsonian Creates First-Ever Digital 3D Presidential Portrait in President Obama appeared first on Architosh.

What and Why

Cal Poly, Pomona professor Michal Fox will be presenting his work with students and space architecture experts in the study and design of NASA’s Deep Space Habit.

The design challenged centered on developing a vertically-oriented habitat for space with volumetric constraints and a “dynamic envelope” which must contain everything its deep space inhabitants rely on from the start, without resupply. Professor Fox and his students used GTeam and Building Information Modeling (BIM) as an approach to setup an inclusive model that automatically managed “consumables” and their locations within the habitat, and compared alternative architectural designs to find the most elegant engineering/architectural solution within a constraint-driven approach.

01 – Upcoming webinar talks about using GTeam for BIM collaboration in study of NASA Deep Space Habitat.

The vertically-oriented deep space habitat was designed, prototyped and tested by architecture students, in collaboration with consultants in space architecture alongside the faculty and students throughout the entire process.

Michael Fox is a founding partner of FoxLin, an Associate Professor of Architecture at Cal Poly Pomona, and is current current President of ACADIA. His practice, teaching and research are centered on interactive, behavioural and kinetic architecture.  He is the author of the book Interactive Architecture and the upcoming book Adaptive: Bio-Robotic Architecture.

02 – Upcoming webinar talks about using GTeam for BIM collaboration in study of NASA Deep Space Habitat.

Michael has won numerous awards in architectural ideas competitions and his work has been featured in international periodicals, books, and exhibitions, most recently at the 2013 Venice Biennale. His work has been funded by NASA, the Annenberg Foundation, the Graham Foundation and others. Michael has taught at MIT, The Hong Polytechnic University, the Art Center College of Design in Pasadena, and SCI-Arc.

Webinar Details

The webinar takes place Tuesday, 11 March 2014, at 12:00 PM PST or 3:00 PM EDT. To register visit this page. The event is free.

The post Event: Collaborative BIM for NASA’s Deep Space Habitat Design appeared first on Architosh.

True UNIX

While Windows-based systems have some tools available that can mimic a Unix-like environment, they often come with limitations to their capabilities, especially when working with large files. Mac OS X however is a true Unix operating system and can handle large files exceptionally well.

01 - HyperWorks now available for Mac OS X. Future iOS apps in the making.

“Much of my work deals with being able to transfer data from very large files into formats that allow further specialized analysis and visualization,” says Mary Leibolt, a structural acoustics analyst at the Naval Warfare Center, Carderock Division. “Having the capability to use the Mac OS X operating system allows me to get my work done on one machine without having to deal with compatibility, transfer media issues, et cetera.”

HyperWorks for Mac OS X offers users more flexibility in choosing which platform to use for their CAE work. Moreover, it opens the world of CAE to a broader range of industries and professions that traditionally work with Mac systems. “With this platform, we extend the availability of HyperWorks to designers and other creative professionals,” says Paolo Sansalvadore, senior vice president of modeling and visualization, Altair, “people who don’t crunch numbers but who interact with the software to visualize their creations and can benefit from the HyperWorks user experience. Altair’s concept and industrial design products, solidThinking Evolve and Inspire, are already available on both Windows and Mac OS X.”

Scientists comprise another important user group that will be able to take advantage of HyperWorks running on the Mac, since much of their work is designed and executed on Mac computers. Last but not least, the new software version also makes CAE more accessible to a broader range of computer-aided design professionals conceiving products largely on Mac systems.

Extending Platforms: iOS and Mac

The extension of HyperWorks to the Mac OS is part of the evolution of Altair software to new devices and platforms. A cloud-based software-as-a-service, HyperWorks On-Demand, is now available through a modern service oriented architecture web-app for running, monitoring and managing simulation workloads and results (Altair Compute Manager). Altair will also offer enhanced mobility through several iOS apps currently under development.

To learn more visit: http://www.altairhyperworks.com/

The post Altair introduces HyperWorks CAE app for Mac OS X appeared first on Architosh.

A quick summary of the latest key elements of the CUDA 4.0 include:

• Nvidia GPUDirect 2.0 Technology — offers support for peer-to-peer communication between GPUs within a single server or workstation. This enables easier and faster multi-GPU programming and application performance
• Unified Virtual Addressing (UVA) — provides a single merged-memory address space for the main system memory and the GPU memories, enabling quicker and easier parallel programming
• Thrust C++ Template Performance Primitives Libraries — provides a collection of powerful open source C++ parallel algorithms and data structures that ease programming for C++ developers. With Thrust, routines such as parallel sorting are 5x to 100x faster

01 - Nvidia's CUDA website is showcasing CUDA 4.0's release.

The new Mac OS X support includes new features in CUDA-GDB. A release candidate for CUDA Toolkit 4.0 will be available free of  charge beginning March 4, 2011. You can enroll in the CUDA developer program here: www.nvidia.com/paralleldeveloper

To learn more about CUDA go here.

The post Nvidia’s new CUDA 4.0 adds Mac OS X support appeared first on Architosh.

Interesting post about Intel’s work on cracking the issue of sharing data between Core and Xeon CPUs and Larrabee GPU co-processors. Future Core and Xeon chips will be able to create a “virtual” shared memory pool that both the CPU and GPU can access so datasets are nto crunched down, serialized, and moved over the slower PCI-Express bus from CPU to GPU, this report discusses.

We know that Apple is working within the OS level to speed up standardized operations so they can be efficiently distributed among available processing cores on both the CPU and GPU levels simultaneously.

There is also a recent post at the same link above about Apple’s OpenCL FFT lib producing very high performance.

The post Tidbits: Apple OpenCL, future GPU/CPU shared memory pools appeared first on Architosh.

OpenCL conformance tests are now available with new conformant products now shipping. Additionally, the OpenCL Working Group membership has expanded to 33 members in the areas of high-performance computing, gaming, middleware, system and silicon vendors. Participating at SC09, an international conference for high-performance computing, Khronos Group members will be highlighting the power and scalability of OpenCL.

OpenCL Products – Trademark Use and Conformance

Before a product can use the OpenCL trademark it must pass a conformance test suite by Khronos for OpenCL 1.0. This suite was released in May 2009 and a number of shipping products have successfully passed including products from AMD, Apple and Nvidia.

In the case of Apple its latest operating system, Snow Leopard, utilizes OpenCL. Apple created OpenCL and opened it up as an industry standard giving control of the standard to the Khronos Group.

“By enabling cross-platform development for heterogeneous architectures, OpenCL is helping to bring GPU compute capability to mainstream applications,” said Patricia Harrell, director of stream computing at AMD.

“Nvidia cares deeply about ensuring that OpenCL developers have the tools they need to easily develop and deploy mainstream applications on more than 150 million Nvidia OpenCL 1.0-capable GPUs,” said Sanford Russell, general manager, GPU Computing software at Nvidia.

The latest list of conforming products for OpenCL 1.0 support can be found here at this web page. So far most products are operating systems and GPUs.

OpenCL Expands Significantly But No Microsoft

With 33 key industry members OpenCL is well on its way to becoming an industry standard as significant as OpenGL–a graphics technology standardized across an array of industry segments and device platforms.

In the gaming industry key members include Activision Blizzard, Electronic Arts and Nvidia and Apple. In terms of silicon systems makers they all the major vendors such as: AMD, ARM, Broadcom, Freescale, IBM, Intel, Nvidia, Samsung, STMicroelectronics, and Texas Instruments.

Key mobile phone makers are also behind OpenCL including: Ericsson, Motorola and Nokia (not to mention Apple). And the US government’s Los Alamos National Laboratory is also a key member.

IBM has released its own OpenCL Development Kit for Linux on Power on AlphaWorks. The company’s Power Architecture processors are used in workstations, servers, mainframe computers and supercomputers. Additionally, IBM is a partner with Sony in Cell processors and is exploring OpenCL on Cell and BE processors as well.

For more information about the Khronos Group’s efforts with OpenCL visit them here.

The post Khronos demonstrates OpenCL momentum at SC09 appeared first on Architosh.

Often many users interested in affordable professional-grade CAD programs on the Mac overlook this French-made program. RealCADD is currently at version 4.07 and one of its primary distinguishing features is that it is produced from the REALBasic programing environment. It is Mac Universal and cross-platform and works also on Windows XP and Linux. RealCADD also supports RBScript which enables the end-user to create their own tools. The Standard version of RealCADD is 95.USD while the Professional version is 135.USD. To learn more visit them here.

CoreMelt GPU Accelerated Plugins for Apple Motion, Final Cut…

CoreMelt released the new V2 Range of GPU accelerated products for image effects for motion artists. There are 180 GPU accelerated plugins that work with Apple’s Final Cut Pro, Final Cut Express, Apple Motion and After Effects CS3/CS4. Running on the Second Generation vTwin FX Engine, these new effects are highly optimized to tap your GPU’s power. To learn more go here.

KnowledgeMinor (yX) for Excel

We were interested to learn about this new Mac product for Mac Excel users. KnowledgeMinor (yX) in partnership with Intel and Microsoft has released a major new data mining application for the Mac. 

KnowledgeMinor (yX) takes advantage of multi-core processors on your Mac and extreme parallelization to achieve data mining speeds more than 600 times faster then previously. Frank Lemke, the creator said, “Speed is always a big factor in data mining. Our previous product KnowledgeMiner was quite remarkable but (yX) for Excel amazed even us with its speed. We want to thank Microsoft and Intel for their help in developing this product”. (yX) is one of very few applications that currently take advantage of multiple cores on the Mac and the only Mac data mining application to do so.

The post Tidbits: RealCADD, CoreMelt, KnowledgeMinor (yX) appeared first on Architosh.

VolView 3.0 for Mac allows medical doctors and researchers 3D volume rendering capabilities with GPU hardware acceleration.

New in this release is a simplified user interface, quick-access toolbar, GPU acceleration under Nvidia graphics platforms, ability to load and display multiple datasets, and a new Bezier contour tool. To learn more visit them at: www.kitware.com

Nvidia Releases OpenGL 3.0 Drivers

Nvidia announced this week that they have released fully supported OpenGL 3.0 drivers for Windows XP, Windows Vista, and Linux platforms on select GeForce and Quadro GPU solutions. 

The company stated: “Further demonstrating our leadership and commitment to OpenGL and the standardization process, these drivers enable software developers to take full advantage of all OpenGL 3.0 capabilities, accelerating their delivery of cutting-edge media platforms and applications. 

Apple’s Mac OS X 10.6 “Snow Leopard” is expected to support OpenGL 3.0 for applications and many of Apple’s professional applications like Motion may push the limits of OpenGL 3.0 capabilities. New OpenGL 3.0 extensions include:

• WGL_create_context
• GLX_create_context
• ARB_draw_instanced
• ARB_geometry_shader4
• ARB_texture_buffer_object

Rhino News/Tutorials

The post Tidbits: VolView 3.0, OpenGL 3.0 and Rhino news appeared first on Architosh.

The reason for NASA’s selection of modo, out of all the 3D rendering design softwares was that they needed precision. Luxology reports that they, NASA designers, relied heavily on modo 302 features such as Micropoly Displacement, which allowed them to get the texture of carbon fiber just right. Luxology is proud to have their softwares participation in helping NASA’s new Moon Rover, which was presented to the public for the first time at President Barack Obama inauguration ceremony yesterday, January 20th 2009. 

modo was used by NASA to develop this new prototype rover

Both parade attendees and TV viewers can view the first public showing of the next-generation SPR, visualized in Luxology’s modo software and produced by NASA for the proposed manned return to the moon in approximately 2020.

For images of the modo-created prototype can be viewed at 

http://www.nasa.gov/exploration/home/LER.html.

The post modo used for Design of NASA’s New Moon Rover! appeared first on Architosh.

Gridgen is Pointwise’s flagship product and the latest release offers improved performance for anisotropic tetrahedral extrusion technique for highly-automated, high-quality, hybrid mesh generation for computational fluid dynamics. 

By optimizing the anisotropic-isotropic blending algorithm and triangle-triangle intersections used for collision detection, the overall mesh generation time was reduced by up to a factor of 7.2. At the same time, a new algorithm for combining tetrahedra into prisms reduces a mesh’s overall cell count by up to 41% over previous releases.

Gridgen works on all major platforms including Windows, Linux, Mac and UNIX. Current support on Mac is limited to PowerPC G5-based computers, running Panther or higher (OS X 10.3+) and requires X11 and an Nvidia GeForce GPU or better. Intel Macs are not yet fully supported but there are directions here on the Pointwise website.

The post PointWise’s Gridgen for Mac – Hybrid Meshing Made Faster appeared first on Architosh.

Apple’s OpenCL Embraced by Industry

Apple developed OpenCL and sponsored its adoption as an industry-open standard. OpenCL (Open Computing Language) greatly improves speed and responsiveness for a wide spectrum of applications in numerous market categories from gaming and entertainment to scientific and medical software. 

Proposed only six months ago, OpenCL was at the heart of Apple’s Snow Leopard operating system preview at its worldwide developer’s conference back in June of 2008. The draft specification became rapidly embraced by the tech world’s biggest names in semiconductors including: IBM, Intel, AMD, Nvidia, ARM, Freescale, Ericsson, Broadcom, Texas Instruments and Samsung. 

Other key players endorsing their OpenCL include gaming software giants EA (Electronic Arts) and Activision Blizzard. Notably absent the from big list of companies embracing OpenCL is Microsoft. However, with both Intel and AMD, the two largest PC chip makers in the industry, supporting it there is a likelihood that Microsoft will eventually add support to OpenCL. 

Microsoft says it has its own proprietary parallel programming technologies in the works. This technology is slated to be apart of DirectX technologies going forward. It should be noted that Microsoft’s Windows-only DirectX technologies are highly adopted in the world of PC gaming but Microsoft has failed to drive OpenGL out of the professional CAD and 3D content creation industries where both UNIX, Linux and Mac OS X are widely deployed and critical for support. Microsoft’s own parallel computing technologies may face a similar fate.

Both OpenGL and OpenCL are managed by the Khronos Group. To learn more visit their site here.

The post Apple’s OpenCL released as ratified 1.0 spec at SIGGRAPH Asia appeared first on Architosh.

Wolfram Research has introduced Mathematica 7, the latest version of its mathematical and computation software program for Mac OS X and other platforms. The new software includes over 500 new features including built-in image processing and analysis. 

Other key improvements include built-in parallel computing, new visualization and graphics capabilities — including comprehensive spline support and the representation of industrial NURBS surfaces.

New support for industrial NURBS

New image processing capabilities.

Mathematica 7 can adds a variety of computable data sources including the complete human genome, weather, astronomical data and GIS and geodesy data. 

Lastly, one key new feature is the new integrated programmatic email features, allowing you to construct any form of email including graphics, sounds and other attachments and to manage recipients and lists programmatically. To learn more click here.

VMware Fusion 2.0.1 Gets Performance Boost

Hot on the heels of Parallel’s new Parallels Desktop 4 for Mac, VMware has updated Fusion to version 2.0.1 adding a new performance boost. The new release will begin to ship Friday and will include a performance boost due to better CPU utilization and the patching of several previously unresolved issues. 

Other improvements focus on better performance when running 3D applications (including games and applications like Google Earth with 3D hardware acceleration) on the latest new MacBook and MacBook Pro and MacBook Air systems. 

Unity view has been improved, there are new hints in the Help menu and improvements to sharing, including nested folders and publishing Windows guest applications to the Mac. Fusion 2 users can download the latest release starting tomorrow here.

The post Tidbits: Mathematica 7 and VM Fusion 2.0.1 appeared first on Architosh.

The Constructech Vision Award has been given to ArchiCAD-based Holden Construction for its use of Building Information Modeling (BIM) software in a large construction project. Holden won the award for its Aquarium Hilton Garden Inn in Atlanta, Georgia. The project demonstrated how a general contractor can use BIM technology to enhance teamwork, build faster, avoid costly change order and diminish the number of field conflicts.

Graphisoft, now a subsidiary of German Nemetschek AG, has recently launched ArchiCAD 12 and introduced the world’s first BIM with multiprocessing support. To learn more about ArchiCAD 12 go here.

Executive Suite Makes Some Changes at Nemetschek North America

Nemetschek North America, the makers of VectorWorks, has announced that Dr. Biplab Sarkar will now be its new Chief Technology Officer (CTO). Dr. Sarkar has been involved with the company for eight years spearheading its advanced 3D modeling technology and came to N.A. from Parametric Technologies.

Eight years ago when Architosh reported this news there was some hope that Dr. Sarkar’s ProE background would foster advanced MCAD on the Mac or perhaps create opportunities for ProE file compatibility between VectorWorks and ProE.

Former CTO Paul Pharr will now become the company’s new Chief Information Officer (CIO) responsible for information management systems and web applications and delivery technologies. These changes are aimed at better positioning the company’s resources.

Nvidia Achieves Top Spot in Distributed Computing Application

Nvidia GPUs are powering over 1 petaflop of processing power to Stanford University’s Folding@Home distributed computing application as of last week, reports the University. Active Nvidia GPUs deliver over 1.25 petaflops or 42% of the total processing power of the application which seeks to understand how proteins affect the human body.

The key to this story however is that Nvidia is achieving this computation contribution via just 11,370 total active processors. In comparison 208,268 CPUs running Windows were active, contributing just 6% of processing power to the project.

Nvidia’s GPUs are delivered in is Nvidia CUDA architecture appliances. “As these statistics show, the impact of NVIDIA GPUs on protein folding simulations has been extraordinary,” said Vijay Pande, associate professor of chemistry, Stanford University and director of the Folding@home project. “Teams that are folding with NVIDIA GPUs are seeing huge boosts to their production and this is helping to accelerate the project significantly.”

To learn more visit this page.

The post Tidbits: ArchiCAD, Vectorworks and Nvidia appeared first on Architosh.

Solid Modeling Solutions provides advanced NURBS-based geometry software libraries sometimes called modeling kernels or engines. The company serves software application developers and one of its best known OEM clients is Nemetschek North America, the makers of VectorWorks. Essentially SMLib is made up of a small elite team of PhD level geniuses in the field of software engineering and mathematics, whose backgrounds include such places as Boeing and leading universities such as MIT. The company was likely at Siggraph for both professional and OEM client reasons. SMLib products are not for end-users but end-users may be interested to learn which modeling kernel technology drives their favorite CAD and 3D applications.

IntegrityWare

Another modeling kernel provider who was at Siggraph wasIntegrityWare. IntegrityWare’s modeling kernel is behind such applications as Autodesk 3ds Max, Viz and Robert McNeel & Associates’ Rhino. Since Rhino is coming to the Mac and there is a current working advanced beta we assume it is based on the same modeling kernel and therefore IntegrityWare’s software libraries for NURBS and polygon modeling are fully OS neutral libraries. Architoshwill confirm this in a later report in September focused on kernel discussion.

RapidMind

Also at Siggraph this year was a company called RapidMind. Another technology company that serves other tech companies rather than end-users,RapidMind first came to our attention at Siggraph Boston two years ago. They were exhibiting in the IBM booth at that time. 

RapidMind and AMD jointly demonstrated real-time medical imagery application at Siggraph this year. RapidMind is the maker of the award-winning RapidMind Multi-Core Development Platform, a series of API’s (application programming interfaces) available to software developers who wish to speed up their applications specifically on multi-core processors.

RapidMind’s API technology can accelerate such things as real-time ray tracing, image reconstruction for the medical industry, intensive image processing in the graphics industry and fluid simulation and computation in gaming, science and animation entertainment industries.

RapidMind API Stack

RapidMind’s API sits as a layer between the underlying hardware and OS and the end-user application. Developers can focus on investing in single-threaded application development and let the RapidMind technology take care of “parallelizing” the application across multiple cores.

The API integrates with C++, requires no new tools, compilers or workflow changes. It leverages AMD and Intel Multi-Core x86 microprocessors and even utilizes GPUs. It also supports the IBM Cell processor used in the IBM Cell Blade server and Sony PlayStation 3 as well as select Cell-based workstations from IBM, Sony and Mercury.

RapidMind runs on Mac OS X 10.5 or better as well as flavors of Windows and Linux. Compatible compilers include Microsoft Visual C++ 7-8, GCC 4 under Linux and Xcode under Mac OS X. A case-study in a QuickTime moviedemonstrates the performance gain using RapidMind.

Most of this page has focused on tools underlying the applications in CAD and 3D that matter to us. This last entry in our Highlights series is from a company behind a lot of products that matter to Mac users.

LightWork Design at Siggraph

LightWork Design is the world leader in rendering software engine technology, used in a vast array of leading Mac CAD and 3D applications. This year the company used Siggraph to introduce LightWorks SE (Studio Edition) and LightWorks 7.9. Key new features in version 7.9 include:

• Ambient Occlusion – adds intelligent predictable ambient light to images for greater depth and realism
• New Complex Reflectance Shaders – targeted at the generation of a wider range of realistic material effects; new color and reflectance shaders include Glossy Reflections and Frosted Glass, Enhanced Car Paint, Plastics and Metallic effects
• CgFX – Extends the number of effects in real-time rendering with external CgFX materials
• Specular Final Gather – refined to concentrate work dependent on the properties of a material: the smoother the surface, the less work will be spent calculating specular Final Gather results, yielding faster rendering times

LightWork Design also demonstrated a new Progressive Rendering technology currently under development. The company is running quarterly rendering contests. See the submissions here.

The post Siggraph Highlights for Mac Users – Part 6 appeared first on Architosh.